Mutations are the cause of many human genetic diseases and the main source of genetic variation in natural populations. Thus, elucidating the mechanisms of mutagenesis is of great significance. The fact that, in mammals, the number of germline cell divisions (and of DNA replications) is higher in males than in females provides an opportunity to test whether mutations result from errors in DNA replication. If this hypothesis is true, we expect higher mutation rate in males than in females (male mutation bias) and higher mutation rate in older males than in younger males (paternal age effect). Here we will employ the tools of comparative genomics and bioinformatics to analyze available mammalian genomic sequences and generate additional data experimentally to test the following specific hypotheses: 1. Errors in DNA replication are the primary sources of insertions and deletions (indels). 2. Nucleotide substitutions, particularly at CpG dinucleotides, depend on the number of germline cell divisions. To test these first two hypotheses we will estimate mutation rates from mammalian whole-genome alignments and compare these rates between sex chromosomes and autosomes. 3. Microsatellite repeat expansions and contractions are caused by errors in DNA replication. This will be tested by observing de novo mutations in single sperm of human males of different ages. 4. The magnitude of male mutation bias and generation time are positively correlated in mammals. To investigate this, we will sequence introns of genes homologous between X and Y in mammals with long generation time (Cetacea and Perissodactyla) and analyze additional data from the literature. The proposed research has direct relevance to issues of public health and clinical genetics. The overwhelming majority of mutations causing human genetic diseases are indels, nucleotide substitutions, and microsatellite repeat expansions/contractions. Moreover, single nucleotide polymorphisms (SNPs), an outcome of nucleotide substitutions, and microsatellites are widely used markers for mapping diseases and traits in association studies. Thus, it is critical to know whether mutations at these loci are driven by replication-dependent or by replication-independent factors (e.g., environmental agents such as free radicals). Additionally, the conclusions of this project will be important for genetic counseling. Namely, our results will indicate whether the age of a father at the time of conception represents a risk factor for pathology in the offspring.